Diversity project tissue cleanup
We need to get this to a usable place. The raw data is scattered over columns in a way similar to how ancestry and tissue were scattered, so this script looks pretty similar.
Cleaning up and pre-processing
We begin by taking a quick look at the tissue type descriptors to get some numbers… Some of these clearly do not contain cell type info, but others do, and sorting those out is joyous.
table(allSRAFinal$Cell_type) # Useful##
## acute myeloblastic leukemia adipose-derived stromal cells (hASCs)
## 6 54
## alveolar basal epithelial cells aortic endothelial cell
## 5 4
## Aortic enothelial cells B cells
## 3 35
## B-Lymphocyte BBB microvascaular
## 30 29
## blood leukocytes breast cancer
## 12 33
## Cardiac fibroblasts cardiomyocyte
## 166 12
## Carotid artery endothelial cells CD138-negative cells
## 6 10
## CD138-positive cells CD14 Primary Cells
## 10 1
## CD14+ peripheral blood cell CD19 Primary Cells
## 4 1
## CD19+ CD3 Primary Cells
## 23 1
## CD34 Primary Cells CD34+ sorted bone marrow progenitors
## 1 12
## CD4 Primary Cells CD4 T cell
## 2 10
## CD4 T cells CD4+ T cells
## 79 42
## CD56 Primary Cells CD56+ peripheral blood cell
## 1 4
## CD8 Primary Cells CD8 T cell
## 2 8
## CD8 T cells CD8+ T cell
## 53 22
## cDC CELL MIXTURE - tissue sample
## 29 1
## classical monocyte CLEAR CELL ADENOCARCINOMA\\, NOS
## 34 1
## cMo cord blood mononuclear cells
## 110 30
## Coronary artery endothelial cells cortical neurons
## 4 4
## endocardial cell Endocardial cells
## 1 3
## ENDOMETRIOID ADENOCARCINOMA endothelial cell of artery (common carotid)
## 1 8
## endothelial cell of great saphenous vein endothelial cell of umbilical artery
## 3 1
## Endothelial cells epithelial
## 48 67
## Epithelial epithelial cells
## 25 4
## epithelial-like epithelial-like cells
## 1 4
## fibroblast Fibroblast
## 38 4
## fibroblasts Firoblast
## 10 22
## Granule Cell of Human Dentate Gyrus great saphenous vein endothelial cell
## 258 1
## Hematopoietic Stem Cells (HSCs) Hepatocyte
## 730 20
## HepLPCs HepLPCs-Hep
## 8 2
## Human Aortic Endothelial Cells Human Brachiocephalic Endothelial Cells
## 6 2
## Human Cardiac Microvascular Endothelial Cells Human Common Carotid Artery
## 1 1
## Human Coronary Artery Endothelial Cells Human Endocardial Cells
## 5 1
## Human primary bronchial epithelial cells Human Pulmonary Artery Endothelial Cells
## 100 4
## Human Umbilical Artery Endothelial Cells Human Umbilical Vein Endothelial Cells
## 2 4
## induced pluripotent stem cell induced pluripotent stem cells
## 332 18
## iPSC iPSC derived cell line
## 45 250
## iPSC derived Motor Neuron (iMN) iPSC-derived cardiomyocytes
## 11 51
## iPSC-derived cortical neuron lung microvascular endothelial cells
## 6 16
## lymphoblast Lymphoblastoid
## 159 38
## M1-polarized HMDM melanocyte
## 48 1
## Microvascular endothelial cell MIXED CLEAR CELL\\, ENDOMETRIOID AND PAPILLARY SEROUS CARCINOMAS
## 4 1
## mononuclear cell monos
## 20 198
## Naive B cell Nasal epithelium
## 21 18
## Natural Killer Cells Neutrophil
## 12 18
## neuts NORMAL TISSUE
## 201 8
## organoid PAPILLARY SEROUS ADENOCARCINOMA
## 39 3
## PAPILLARY SEROUS AND ENDOMETRIOID ADENOCARCINOMA PAPILLARY SEROUS CARCINOMA
## 1 8
## PAPILLARY SEROUS CYSTADENOCARCINOMA PBMC
## 4 209
## PBMC derived macrophage (HMDM) PBMCs
## 48 12
## PD1 pDC
## 74 30
## peripheral blood mononuclear cells Peripheral blood mononuclear cells
## 24 54
## Platelet PMN
## 99 61
## primary primary cell
## 16 12
## Primary cell Primary hepatocytes
## 32 2
## Primary vascular smooth muscle cells prostate
## 8 12
## prostate cancer Pulmonary artery endothelial cells
## 24 2
## Renal artery endothelial cells SEROUS ADENOCARCINOMA
## 7 2
## SEROUS CARCINOMA SEROUS SURFACE PAPILLARY CARCINOMA
## 4 3
## Small Airway Epithelial Cells (SAECs) small airway epithelium (SAE)
## 36 12
## Stem Cell superior vena cava endothelial cells
## 8 1
## T cell T cells
## 18 33
## Tfh Treg
## 74 74
## Umbilical artery endothelial cells Umbilical vein endothelial cells
## 2 3
## White blood cells Whole blood
## 67 28table(allSRAFinal$Tissue) # Useful##
## AC ACD plasma
## 18 12
## Adenocarcinoma of the kidney Adenocarcinoma of the large intestine
## 1 2
## Adenocarcinoma of the lung Adenocarcinoma\\, invasive
## 2 2
## Adenocarcinoma\\, invasive of the lung Adenocarcinoma\\, mixed subtype
## 1 1
## Adenocarcinoma\\, papillary predominant adipose
## 1 50
## Adjacent normal Adjacent normal tissue
## 4 14
## Adrenal adult cartilage
## 1 3
## Aorta Apical left ventricle
## 51 32
## Atrial Tissue autonomic_ganglia
## 24 16
## Base of Tongue biliary_tract
## 10 8
## BLA bladder
## 60 1
## Bladder cancer blood
## 5 4873
## Blood blood plasma enriched for extracellular vesicles
## 1287 82
## bone bone marrow
## 26 33
## Bone marrow Bone Marrow
## 47 42
## brain Brain
## 288 135
## Brain – choroid plexus Brain (DLPFC)
## 15 32
## Brain capillaries breast
## 29 137
## Breast breast cancer tissue
## 9 82
## Brodmann Area 38 Cancer tissue
## 53 14
## Carotid body tumor Caudate
## 24 131
## CE cecum
## 60 2
## central_nervous_system CHD whole blood 1
## 64 1
## CHD whole blood 2 CHD whole blood 3
## 1 1
## CHD whole blood 4 CHD whole blood 5
## 1 1
## CHD whole blood 6 Clear cell adenocarcinoma of the kidney
## 1 3
## colon Colon biopsy
## 132 16
## colorectal cancer Common carotid artery
## 39 6
## Connective Connective tissue
## 22 18
## Coronary artery DC
## 4 4
## distal ileum DLPFC
## 1 23
## Dorsolateral prefrontal cortex EDTA plasma
## 80 12
## Endocardium endometrial tissue
## 3 10
## endometrium F
## 28 6
## FALLOPIAN TUBE Fat
## 8 54
## femoral head fibroblast
## 4 8
## Floor of Mouth foetal lung
## 1 12
## Foreskin Gastric (Benign)
## 4 2
## Gastric (Normal) Gastric (Tumor)
## 3 36
## gingival Great Saphenous vein
## 12 3
## haematopoietic_and_lymphoid_tissue HC whole blood 1
## 178 1
## HC whole blood 2 HC whole blood 3
## 1 1
## HC whole blood 4 HC whole blood 5
## 1 1
## HC whole blood 6 Heart\\, left ventricle
## 1 1
## Hematopoietic tissue Heparin plasma
## 20 44
## Hepatocellular carcinoma of the liver Hepatocellular carcinoma\\, NOS
## 1 2
## HF http://purl.obolibrary.org/obo/BTO_0000784
## 2 20
## Human Hippocampus Hyppocampus
## 258 12
## ICV Ile
## 59 2
## Ileal biopsy Ileal pouch
## 245 39
## Ileum Induced pluripotent stem cell line derived from adult skin (leg) fibroblasts
## 195 18
## induced pluripotent stem cell-derived cortical neurons iPSC
## 22 24
## iPSC_CRISPR_correcte-derived cortical neuron\\, replicate 1 iPSC_CRISPR_correcte-derived cortical neuron\\, replicate 2
## 1 1
## iPSC_CRISPR_correcte-derived cortical neuron\\, replicate 3 iPSC_CRISPR_corrected\\, replicate 1
## 1 1
## iPSC_CRISPR_corrected\\, replicate 2 iPSC_CRISPR_corrected\\, replicate 3
## 1 1
## iPSC-derived cells iPSC-derived cortical neuron\\, replicate 1
## 304 1
## iPSC-derived cortical neuron\\, replicate 2 iPSC-derived cortical neuron\\, replicate 3
## 1 1
## iPSC\\, replicate 1 iPSC\\, replicate 2
## 1 1
## iPSC\\, replicate 3 kidney
## 1 33
## Kidney cortex Kidney medulla
## 1 1
## large_intestine Larynx
## 58 1
## Left Atrium Left Lung
## 8 60
## Left Ventricle leukapheresis
## 8 1
## liver Liver
## 71 229
## Liver\\, Epithelial LN
## 12 66
## lung Lung
## 198 67
## Lung (Endotracheal aspirate\\, ETA) lung; derived from metastatic site: lymph node
## 4 6
## lymph node Lymphocytes
## 2 18
## M mammary gland/breast cancer; derived from metastatic site: pleural effusion
## 2 21
## mammary gland/breast; derived from metastatic site: pleural effusion mammary gland\\, breast; derived from metastatic site: pleural effusion
## 1 24
## Maternal Whole Blood mediastinum
## 238 2
## melanoma metastatic ovarian cancer
## 1 129
## missing myocardium (atrial biopsy)
## 4 53
## Nac NAC
## 130 58
## Nasal epithelium Nasal olfactory mucosa: middle turbinate
## 18 17
## Nasal RNA Normal bladder epithelium
## 65 5
## normal breast tissue normal prostate
## 5 16
## oesophagus OMENTUM
## 25 9
## Organotypic raft cultures (ORC) containing keratinocytes ovarian clear cell carcinoma
## 23 9
## ovary Ovary
## 48 10
## OVARY ovary cystadenocarcinoma
## 10 9
## OVARY TUMOR ovary: ascites
## 8 21
## pancreas pancreas/duct
## 71 16
## parathyroid PBMC
## 15 197
## peripheral blood Peripheral blood
## 771 194
## Peripheral blood mononuclear cells Peripheral Retina
## 20 16
## Peripheral RPE-Choroid-Sclera Peripheral venous blood
## 15 20
## PERITONEUM PFC
## 1 60
## pheripheral blood Placenta
## 18 32
## Plasma exosome pleura
## 6 11
## Postmortem brain\\, dorsolateral prefrontal cortex (DLPFC) Postmortem Cervical Spinal Section Homogenate
## 52 15
## Precuneus prefrontal cortex
## 10 36
## Prepouch ileum primary breast tumour
## 36 18
## Primary Heart Tissue Primary nasal epithelial cells
## 22 8
## primary skin fibroblasts primary tumor
## 24 36
## primary tumor tissue from high grade serous ovarian tumors Primary visual cortex
## 72 10
## prostate prostate cancer
## 83 114
## prostate gland (matched normal tissue) prostate tumor
## 28 16
## Pterygia Pulmonary artery
## 12 2
## Purified platelets Putamen
## 1 131
## rec Rec
## 2 26
## rectum Renal artery
## 76 7
## Renal cell carcinoma retina
## 20 16
## Right Atrium Right Lung
## 8 40
## salivary_gland SC
## 2 9
## Serum sigmoid colon
## 12 25
## Skeletal muscle skin
## 16 182
## Skin Skin biopsies
## 83 36
## skin tissue small_intestine
## 6 1
## soft_tissue Stem cell population
## 19 8
## stomach Substanstia nigra
## 63 34
## TC testes
## 2 1
## testis Testis
## 1 14
## thyroid Thyroid
## 12 19
## TI Tongue
## 2 6
## Tonsil Trachea/Bronchus
## 6 12
## triple negative breast cancer tumor (FFPE) tumor
## 42 15
## Tumor Tumor tissue
## 4 15
## Tumor\\, unspecified of the large intestine Umbilical artery
## 1 2
## Umbilical vein upper_aerodigestive_tract
## 3 32
## urinary_tract uterine clear cell carcinoma
## 26 4
## uterus Uterus
## 36 2
## vaginal vastus lateralis muscle
## 16 132
## vastus lateralis skeletal muscle vessel
## 671 20
## whole blood Whole blood
## 445 348
## Whole Blood Whole skin
## 56 16table(allSRAFinal$tissue_type) # Useful##
## Adipose Adipose Tissue Adrenal Gland Aorta Bladder
## 2 1 2 2 1
## Esophagus Gastric Left Ventricle Liver Lung
## 2 3 2 1 2
## Ovary Pancreas Psoas Muscle Right Atrium Right Ventricle
## 1 2 3 1 2
## Sigmoid Colon Small Intestine Spleen the endothelium of blood vessels Thymus
## 3 3 3 26 1table(allSRAFinal$Sample_type) # Somewhat useful##
## blood tissue sample cell culture Cell culture Cell Culture Transcriptome Sequence cortical neurons
## 16 1036 24 29 4
## frozen hiPSC-derived neurons leukocyte liquid cell culture Plasma sample\\, platelet-poor
## 3 28 12 8 68
## Primary Cell Serum sample the cirrhotic the liver tissues Tissue
## 42 12 9 9 6
## tissue sample
## 15table(allSRAFinal$Sample.Name) # Nope## < table of extent 0 >table(allSRAFinal$source_name) # Useful##
## 12-cell blastomere
## 6
## 16-cell blastomere
## 7
## 4 mm punch biopsy
## 16
## 4-cell blastomere
## 5
## 8-cell blastomere
## 23
## Active_Baseline
## 29
## Active_M30
## 2
## Active_Wk 104
## 18
## Active_Wk 104 (130)
## 1
## Active_Wk 104 (or M30)
## 1
## Active_Wk 26
## 28
## Active_Wk 52
## 30
## Active_Wk 78
## 26
## adipose
## 50
## Adjacent normal
## 4
## Aged A_peripheral blood HSC
## 94
## Aged B_peripheral blood HSC
## 83
## Aged C_peripheral blood HSC
## 94
## Aged D_peripheral blood HSC
## 94
## Air Liquid Interface culture
## 36
## AnCg_Bipolar Disorder
## 24
## AnCg_Control
## 24
## AnCg_Major Depression
## 24
## AnCg_Schizophrenia
## 24
## articular cartilage of the femoral condyle and/or tibial plateau
## 3
## ASC_Pop2
## 11
## ASC_Pop3
## 11
## ASC_PopA
## 10
## ASC_PopB
## 11
## ASC_PopD
## 10
## B
## 18
## B-Lymphocyte
## 108
## Bcell
## 28
## blood
## 350
## Blood
## 1386
## Blood\\, case
## 110
## Blood\\, control
## 245
## bone tissue cells
## 4
## Brain
## 89
## Breast
## 196
## Breast tissue
## 60
## Brodmann Area 38
## 53
## bronchoalveolar lavage
## 100
## Caudate
## 72
## CD4
## 56
## CD8
## 54
## CD8+ sorted T-cells
## 63
## CKD_vastus lateralis muscle
## 6
## Clear cell tumor
## 13
## Colon
## 53
## control
## 8
## Control_Ileum
## 37
## Control_vastus lateralis muscle
## 3
## Control\\, classical monocyte
## 18
## coronary artery calcification case
## 8
## COVID-19 blood
## 4
## COVID-19 TA
## 4
## Crohn's disease_Ileal biopsy
## 210
## Crohn's Disease_Ileum
## 112
## Dialysis_vastus lateralis muscle
## 4
## DLPFC_Bipolar Disorder
## 23
## DLPFC_Control
## 24
## DLPFC_Major Depression
## 23
## DLPFC_Schizophrenia
## 24
## Donor_1 HepLPCs-Hep
## 1
## Donor_1 PHCs
## 1
## Donor_1 TEM Day10
## 1
## Donor_1 TEM Day4
## 1
## Donor_1 TEM P10
## 1
## Donor_1 TEM P5
## 1
## Donor_2 PHCs
## 1
## Donor_2 TEM Day10
## 1
## Donor_2 TEM Day4
## 1
## Donor_2 TEM P10
## 1
## Donor_2 TEM P5
## 1
## Donor_3 HepLPCs-Hep
## 1
## Duodenum
## 32
## FALLOPIAN TUBE
## 8
## Fibroblast
## 32
## fibroblast derived iPSCs
## 6
## frontal cortex of the human brain
## 34
## full-thickness sigmoid colon tissue
## 25
## Glioblastoma
## 12
## HC blood
## 17
## Healthy control
## 11
## Healthy control blood
## 6
## Healthy donor CD34+ Bone Marrow
## 12
## healthy skin
## 1
## Heart
## 71
## hematoma
## 178
## Huh-7 Cell Line
## 12
## HuH7 1D142
## 3
## Human Aortic smooth muscle cells
## 8
## Human colorectal adenocarcinoma
## 6
## human Dorsal Root Ganglia (hDRG)
## 1
## human induced pluripotent stem cells
## 317
## human omental metastasis
## 129
## Human postmortem brain sample
## 34
## Human skin microvascular endothelium
## 4
## IBS_Ileum
## 1
## Ideopathic pulmonary disease\\, PBMCs
## 54
## Ileum
## 47
## Ileum1
## 1
## Ileum2
## 1
## In vitro differentiated plasma cells
## 53
## induced pluripotent stem cell-derived cardiomyocytes
## 44
## induced pluripotent stem cell-derived cortical neurons
## 22
## induced pluripotent stem cells
## 18
## inner cell mass
## 13
## Insulin resistant_iPSC_no treament
## 6
## Insulin resistant_iPSC_treated
## 6
## Insulin sensitive_iPSC_no treament
## 6
## Insulin sensitive_iPSC_treated
## 6
## iPSC
## 11
## iPSC-derived cardiomyocytes day 15
## 16
## iPSC-derived cardiomyocytes day 27 Not T3 treated
## 8
## iPSC-derived cardiomyocytes day 27 T3 treated
## 27
## keloid
## 4
## Left Lung
## 60
## Left Ventricle
## 361
## Left Ventricle Endocardium
## 11
## Leicester_Active_TB
## 53
## Leicester_Control
## 50
## Leicester_LTBI
## 49
## Liver
## 16
## LN
## 66
## Longitudnal_Leicester_Control_Non_progressor
## 69
## Longitudnal_Leicester_LTBI_Non_progressor
## 69
## Longitudnal_Leicester_LTBI_Progressor
## 23
## Lower leg skin
## 20
## Lung
## 66
## lung microvascular endothelial cell
## 16
## Lymphocytes
## 55
## Macrophage
## 96
## Maternal Whole Blood_Chinese
## 146
## Maternal Whole Blood_Indian
## 44
## Maternal Whole Blood_Malay
## 48
## metastatic tumor
## 11
## monocyte
## 28
## Monocytes
## 18
## Muscle (ileum)
## 3
## Muscle-Colon
## 3
## Muscle-Duod
## 3
## Muscle-Ileum
## 3
## Nac
## 71
## NAc\\, Caudate\\, and Putamen_pair1
## 3
## NAc\\, Caudate\\, and Putamen_pair103
## 3
## NAc\\, Caudate\\, and Putamen_pair105
## 3
## NAc\\, Caudate\\, and Putamen_pair107
## 3
## NAc\\, Caudate\\, and Putamen_pair109
## 3
## NAc\\, Caudate\\, and Putamen_pair11
## 3
## NAc\\, Caudate\\, and Putamen_pair111
## 3
## NAc\\, Caudate\\, and Putamen_pair115
## 3
## NAc\\, Caudate\\, and Putamen_pair117
## 3
## NAc\\, Caudate\\, and Putamen_pair119
## 3
## NAc\\, Caudate\\, and Putamen_pair129
## 3
## NAc\\, Caudate\\, and Putamen_pair133
## 3
## NAc\\, Caudate\\, and Putamen_pair139
## 3
## NAc\\, Caudate\\, and Putamen_pair145
## 3
## NAc\\, Caudate\\, and Putamen_pair149
## 3
## NAc\\, Caudate\\, and Putamen_pair15
## 3
## NAc\\, Caudate\\, and Putamen_pair153
## 3
## NAc\\, Caudate\\, and Putamen_pair157
## 3
## NAc\\, Caudate\\, and Putamen_pair159
## 3
## NAc\\, Caudate\\, and Putamen_pair163
## 3
## NAc\\, Caudate\\, and Putamen_pair167
## 3
## NAc\\, Caudate\\, and Putamen_pair171
## 3
## NAc\\, Caudate\\, and Putamen_pair173
## 3
## NAc\\, Caudate\\, and Putamen_pair177
## 3
## NAc\\, Caudate\\, and Putamen_pair183
## 3
## NAc\\, Caudate\\, and Putamen_pair187
## 3
## NAc\\, Caudate\\, and Putamen_pair19
## 3
## NAc\\, Caudate\\, and Putamen_pair193
## 3
## NAc\\, Caudate\\, and Putamen_pair195
## 3
## NAc\\, Caudate\\, and Putamen_pair197
## 3
## NAc\\, Caudate\\, and Putamen_pair201
## 3
## NAc\\, Caudate\\, and Putamen_pair205
## 3
## NAc\\, Caudate\\, and Putamen_pair207
## 3
## NAc\\, Caudate\\, and Putamen_pair21
## 3
## NAc\\, Caudate\\, and Putamen_pair211
## 3
## NAc\\, Caudate\\, and Putamen_pair213
## 3
## NAc\\, Caudate\\, and Putamen_pair217
## 3
## NAc\\, Caudate\\, and Putamen_pair219
## 3
## NAc\\, Caudate\\, and Putamen_pair25
## 3
## NAc\\, Caudate\\, and Putamen_pair27
## 3
## NAc\\, Caudate\\, and Putamen_pair29
## 3
## NAc\\, Caudate\\, and Putamen_pair37
## 3
## NAc\\, Caudate\\, and Putamen_pair39
## 3
## NAc\\, Caudate\\, and Putamen_pair41
## 3
## NAc\\, Caudate\\, and Putamen_pair43
## 3
## NAc\\, Caudate\\, and Putamen_pair53
## 3
## NAc\\, Caudate\\, and Putamen_pair59
## 3
## NAc\\, Caudate\\, and Putamen_pair63
## 3
## NAc\\, Caudate\\, and Putamen_pair65
## 3
## NAc\\, Caudate\\, and Putamen_pair67
## 3
## NAc\\, Caudate\\, and Putamen_pair69
## 3
## NAc\\, Caudate\\, and Putamen_pair73
## 3
## NAc\\, Caudate\\, and Putamen_pair77
## 3
## NAc\\, Caudate\\, and Putamen_pair81
## 3
## NAc\\, Caudate\\, and Putamen_pair83
## 3
## NAc\\, Caudate\\, and Putamen_pair87
## 3
## NAc\\, Caudate\\, and Putamen_pair89
## 3
## NAc\\, Caudate\\, and Putamen_pair95
## 3
## NAc\\, Caudate\\, and Putamen_pair99
## 3
## nAcc_Bipolar Disorder
## 24
## nAcc_Control
## 22
## nAcc_Major Depression
## 22
## nAcc_Schizophrenia
## 23
## Naïve B cell
## 21
## Nasal RNA
## 65
## nasal_tissue
## 6
## Neurons isolated from post-mortem dlPFC
## 36
## Neutrophil
## 18
## Non-IBD control_Ileal biopsy
## 35
## normal scar
## 3
## normal tissue matched to tumor 1
## 1
## normal tissue matched to tumor 10
## 1
## normal tissue matched to tumor 11
## 1
## normal tissue matched to tumor 12
## 1
## normal tissue matched to tumor 13
## 1
## normal tissue matched to tumor 14
## 1
## normal tissue matched to tumor 15
## 1
## normal tissue matched to tumor 16
## 1
## normal tissue matched to tumor 2
## 1
## normal tissue matched to tumor 3
## 1
## normal tissue matched to tumor 4
## 1
## normal tissue matched to tumor 5
## 1
## normal tissue matched to tumor 6
## 1
## normal tissue matched to tumor 7
## 1
## normal tissue matched to tumor 8
## 1
## normal tissue matched to tumor 9
## 1
## normal-weight child with asthma_CD4+ T cells
## 21
## obese child with asthma_CD4+ T cells
## 21
## OMENTUM
## 9
## oral mucosa
## 66
## Organotypic raft cultures
## 23
## OVARY
## 10
## OVARY TUMOR
## 8
## Overy
## 10
## P427.5
## 2
## P427.6
## 1
## P427.7
## 2
## P500.12
## 2
## P500.5
## 2
## P500.8
## 2
## P501.5
## 2
## P501.6
## 2
## P501.7
## 2
## P502.12
## 3
## P502.4
## 2
## P502.8
## 2
## P503.5
## 1
## P503.6
## 2
## P503.7
## 2
## P505.5
## 1
## P505.6
## 2
## P505.7
## 2
## P506.5
## 2
## P506.6
## 1
## P506.7
## 2
## P507.5
## 2
## P508.5
## 2
## P508.6
## 2
## P509.5
## 2
## P510.5
## 2
## P511.5
## 2
## P511.6
## 2
## P512.5
## 1
## P513.5
## 2
## P600.5
## 2
## P600.6
## 2
## P601
## 2
## P602
## 4
## P603
## 2
## P604.6
## 2
## P604.7
## 1
## P607.5
## 2
## P701
## 2
## PAXgene blood
## 1857
## PBMC
## 346
## PBMCs
## 280
## Peripheral blood leukocytes
## 8
## peripheral blood mononuclear cells
## 24
## Peripheral Retina
## 16
## Peripheral RPE-Choroid-Sclera
## 15
## Peripheral venous blood
## 20
## PERITONEUM
## 1
## Pituitary gland
## 7
## Placebo_Baseline
## 13
## Placebo_M30
## 2
## Placebo_Wk 104
## 8
## Placebo_Wk 26
## 10
## Placebo_Wk 52
## 12
## Placebo_Wk 78
## 15
## Post-mortem dosolateral prefrontal cortex
## 34
## Postmortem Brain\\, DLPFC
## 52
## Primary Heart Tissue
## 22
## primary ovarian tumor tissue
## 72
## primary tumor
## 550
## Prostate Core Biopsy
## 32
## Pterygium
## 12
## Pulmonary sarcoidosis
## 8
## Putamen
## 72
## Rectum
## 2
## retina
## 8
## Retina
## 8
## right atrial appendage biopsies
## 166
## Right Lung
## 40
## RNA from cardiac left ventricle
## 32
## RNA from white blood cells
## 67
## RNA-seq
## 25
## sensory neurons induced by long term NGN2-BRN3A expression in iPSC-derived neural crest (NC_iSN1)
## 3
## sensory neurons induced by NGN2-BRN3A expression in iPSCs (iSN)
## 3
## sensory neurons induced by short term NGN2-BRN3A expression in iPSC-derived nueral crest (NC_iSN2)
## 3
## Skin
## 48
## Skin biopsy
## 13
## skin fibroblasts from biopsies
## 5
## small airway epithelium brushing
## 12
## SPTB female
## 8
## SPTB male
## 8
## SSc\\, classical monocyte
## 16
## Synovial Tissue
## 92
## Term female
## 8
## Term male
## 8
## Tissue
## 53
## triple negative breast cancer tumor (FFPE)
## 42
## trophectoderm
## 9
## tumor
## 1
## Tumor
## 3
## tumor 1
## 1
## tumor 10
## 1
## tumor 11
## 1
## tumor 12
## 1
## tumor 13
## 1
## tumor 14
## 1
## tumor 15
## 1
## tumor 16
## 1
## tumor 2
## 1
## tumor 3
## 1
## tumor 4
## 1
## tumor 5
## 1
## tumor 6
## 1
## tumor 7
## 1
## tumor 8
## 1
## tumor 9
## 1
## tumor adjacent normal
## 1
## U3A cell line
## 18
## Ulcerative Colitis_Ileum
## 44
## Unclassified IBD_Ileum
## 1
## undifferentiated induced pluripotent stem cells (iPSC)
## 3
## unknown primary or metastatic tumor
## 13
## Urothelial cancer FFPE tissue
## 52
## Uterus
## 35
## vastus lateralis muscle
## 119
## Vena femoralis left
## 13
## Vena femoralis right
## 4
## Vena femoralis Right
## 16
## Vena sbclavian right
## 8
## Vena subclavian left
## 1
## Vena Subclavian left
## 8
## Vena subclavian right
## 3
## Vena Subclavian Right
## 1
## whole blood
## 1028
## Whole blood
## 253
## Whole Blood
## 113
## whole blood leukocyte
## 37
## whole blood sample\\, healthy control
## 45
## whole blood sample\\, systemic sclerosis patient
## 232
## Whole blood_activity_tertile_one
## 7
## Whole blood_activity_tertile_three
## 5
## Whole blood_activity_tertile_two
## 9
## Yg A_peripheral blood HSC
## 94
## Yg B_peripheral blood HSC
## 83
## Yg C_peripheral blood HSC
## 94
## Yg D_peripheral blood HSC
## 94table(allSRAFinal$source_material_cell_type) # goes with the next column##
## Fibroblast
## 332table(allSRAFinal$source_material) ###
## Skin tissue
## 332table(allSRAFinal$SAMPLE_TYPE) # Somewhat useful##
## 12-cell blastomere 16-cell blastomere 4-cell blastomere
## 6 7 5
## 8-cell blastomere cell cell culture
## 23 12 190
## Cell Culture Cell culture Biological Replica 1 Cell culture Biological Replica 2
## 49 7 7
## Cell culture Biological Replica 3 Day 0 Biological Replicate A WTC-11 Day 0 Biological Replicate B WTC-11
## 7 1 1
## Day 0 Biological Replicate C WTC-11 Day 13 Biological Replicate A WTC-11 Day 13 Biological Replicate B WTC-11
## 1 1 1
## Day 13 Biological Replicate C WTC-11 Day 18 Biological Replicate A WTC-11 Day 18 Biological Replicate B WTC-11
## 1 1 1
## Day 18 Biological Replicate C WTC-11 Day 28 Biological Replicate A WTC-11 Day 28 Biological Replicate B WTC-11
## 1 1 1
## Day 28 Biological Replicate C WTC-11 Day 7 Biological Replicate A WTC-11 Day 7 Biological Replicate B WTC-11
## 1 1 1
## Day 7 Biological Replicate C WTC-11 DNA-seq and RNA-seq FFPE
## 1 16 17
## inner cell mass iPSC- derived cerebral organoids iPSC-derived cell types
## 13 24 304
## Laser Capture Microdissection followed by RNA-seq leukocyte organoid
## 258 32 39
## PBMC PBMC from blood polysomal RNA
## 5 36 6
## Post-mortem Resected RNA
## 20 30 82
## RNA-seq RNASeq single cell
## 8 1 47
## Tissue tissue cytological samples tissue sample
## 38 17 1040
## total RNA trophectoderm whole blood
## 6 9 60
## Whole Blood
## 56table(allSRAFinal$Sampling_site) # A single study once again...##
## adrenal gland bile duct bladder bone brain breast chest wall colon endometrium kidney liver lung
## 1 2 16 1 1 15 3 1 1 1 4 33
## lymph node muscle omentum other ovary pancreas pleural cavity skin soft tissue
## 11 1 2 1 1 3 2 1 4# The most useful ones are gonna be Tissue, Cell_type and source_name, I think
(table(is.na(allSRAFinal$Tissue), is.na(allSRAFinal$Cell_type)))##
## FALSE TRUE
## FALSE 2568 13886
## TRUE 2333 6601head(allSRAFinal[!is.na(allSRAFinal$Tissue) & !is.na(allSRAFinal$Cell_type), ])## SRA.Study Center.Name BioSample Organism LibrarySource Cell_type Tissue Cell_line disease ETHNICITY disease_state Diagnosis tissue_type clinical_information
## 58488 SRP166898 GEO SAMN10315825 Homo sapiens TRANSCRIPTOMIC Tfh PBMC <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 58489 SRP166898 GEO SAMN10315821 Homo sapiens TRANSCRIPTOMIC Tfh PBMC <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 58490 SRP166898 GEO SAMN10315820 Homo sapiens TRANSCRIPTOMIC Tfh PBMC <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 58491 SRP166898 GEO SAMN10315934 Homo sapiens TRANSCRIPTOMIC Tfh PBMC <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 58492 SRP166898 GEO SAMN10315933 Homo sapiens TRANSCRIPTOMIC Tfh PBMC <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 58493 SRP166898 GEO SAMN10315931 Homo sapiens TRANSCRIPTOMIC Tfh PBMC <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## RACE health_state Population DONOR_HEALTH_STATUS DONOR_ETHNICITY reported_race ReleaseDate create_date ancestry donor_id Sample_type source_name
## 58488 White <NA> <NA> <NA> <NA> <NA> 2018-12-27 2018-10-25 17:04:00 <NA> <NA> <NA> PBMC
## 58489 White <NA> <NA> <NA> <NA> <NA> 2018-12-27 2018-10-25 17:04:00 <NA> <NA> <NA> PBMC
## 58490 White <NA> <NA> <NA> <NA> <NA> 2018-12-27 2018-10-25 17:04:00 <NA> <NA> <NA> PBMC
## 58491 Black or African American <NA> <NA> <NA> <NA> <NA> 2018-12-27 2018-10-25 17:04:00 <NA> <NA> <NA> PBMC
## 58492 Black or African American <NA> <NA> <NA> <NA> <NA> 2018-12-27 2018-10-25 17:04:00 <NA> <NA> <NA> PBMC
## 58493 White <NA> <NA> <NA> <NA> <NA> 2018-12-27 2018-10-25 17:04:00 <NA> <NA> <NA> PBMC
## source_material_cell_type source_material Organism_part race.ethnicity primary_race SAMPLE_TYPE Sampling_site finalRace finalGeography hispanic
## 58488 <NA> <NA> <NA> <NA> <NA> <NA> <NA> White <NA> <NA>
## 58489 <NA> <NA> <NA> <NA> <NA> <NA> <NA> White <NA> <NA>
## 58490 <NA> <NA> <NA> <NA> <NA> <NA> <NA> White <NA> <NA>
## 58491 <NA> <NA> <NA> <NA> <NA> <NA> <NA> Black or African American <NA> <NA>
## 58492 <NA> <NA> <NA> <NA> <NA> <NA> <NA> Black or African American <NA> <NA>
## 58493 <NA> <NA> <NA> <NA> <NA> <NA> <NA> White <NA> <NA>(table(is.na(allSRAFinal$Tissue), is.na(allSRAFinal$source_name)))##
## FALSE TRUE
## FALSE 7303 9151
## TRUE 7167 1767head(allSRAFinal[!is.na(allSRAFinal$Tissue) & !is.na(allSRAFinal$source_name), ])## SRA.Study Center.Name BioSample Organism LibrarySource Cell_type Tissue Cell_line disease ETHNICITY disease_state Diagnosis tissue_type clinical_information
## 48644 SRP043166 GEO SAMN02852956 Homo sapiens TRANSCRIPTOMIC <NA> Pterygia <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 48645 SRP043166 GEO SAMN02852957 Homo sapiens TRANSCRIPTOMIC <NA> Pterygia <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 48646 SRP043166 GEO SAMN02852961 Homo sapiens TRANSCRIPTOMIC <NA> Pterygia <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 48647 SRP043166 GEO SAMN02852955 Homo sapiens TRANSCRIPTOMIC <NA> Pterygia <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 48648 SRP043166 GEO SAMN02852959 Homo sapiens TRANSCRIPTOMIC <NA> Pterygia <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 48649 SRP043166 GEO SAMN02852964 Homo sapiens TRANSCRIPTOMIC <NA> Pterygia <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## RACE health_state Population DONOR_HEALTH_STATUS DONOR_ETHNICITY reported_race ReleaseDate create_date ancestry donor_id Sample_type source_name
## 48644 Caucasian <NA> <NA> <NA> <NA> <NA> 2015-07-22 2015-12-09 02:16:00 <NA> <NA> <NA> Pterygium
## 48645 Caucasian <NA> <NA> <NA> <NA> <NA> 2015-07-22 2015-12-09 02:16:00 <NA> <NA> <NA> Pterygium
## 48646 Caucasian <NA> <NA> <NA> <NA> <NA> 2015-07-22 2015-12-09 02:16:00 <NA> <NA> <NA> Pterygium
## 48647 Caucasian <NA> <NA> <NA> <NA> <NA> 2015-07-22 2015-12-09 02:15:00 <NA> <NA> <NA> Pterygium
## 48648 Caucasian <NA> <NA> <NA> <NA> <NA> 2015-07-22 2015-12-09 02:16:00 <NA> <NA> <NA> Pterygium
## 48649 Caucasian <NA> <NA> <NA> <NA> <NA> 2015-07-22 2015-12-09 01:31:00 <NA> <NA> <NA> Pterygium
## source_material_cell_type source_material Organism_part race.ethnicity primary_race SAMPLE_TYPE Sampling_site finalRace finalGeography hispanic
## 48644 <NA> <NA> <NA> <NA> <NA> <NA> <NA> White <NA> <NA>
## 48645 <NA> <NA> <NA> <NA> <NA> <NA> <NA> White <NA> <NA>
## 48646 <NA> <NA> <NA> <NA> <NA> <NA> <NA> White <NA> <NA>
## 48647 <NA> <NA> <NA> <NA> <NA> <NA> <NA> White <NA> <NA>
## 48648 <NA> <NA> <NA> <NA> <NA> <NA> <NA> White <NA> <NA>
## 48649 <NA> <NA> <NA> <NA> <NA> <NA> <NA> White <NA> <NA>(table(is.na(allSRAFinal$Cell_type), is.na(allSRAFinal$source_name)))##
## FALSE TRUE
## FALSE 2824 2077
## TRUE 11646 8841head(allSRAFinal[!is.na(allSRAFinal$Cell_type) & !is.na(allSRAFinal$source_name), ])## SRA.Study Center.Name BioSample Organism LibrarySource Cell_type Tissue Cell_line disease ETHNICITY disease_state Diagnosis tissue_type
## 48582 SRP120018 GEO SAMN07789039 Homo sapiens TRANSCRIPTOMIC HepLPCs <NA> <NA> <NA> Asian <NA> <NA> <NA>
## 48583 SRP120018 GEO SAMN07789036 Homo sapiens TRANSCRIPTOMIC Primary hepatocytes <NA> <NA> <NA> Asian <NA> <NA> <NA>
## 48656 SRP120018 GEO SAMN07789049 Homo sapiens TRANSCRIPTOMIC HepLPCs-Hep <NA> <NA> <NA> Asian <NA> <NA> <NA>
## 48657 SRP120018 GEO SAMN07789046 Homo sapiens TRANSCRIPTOMIC HepLPCs <NA> <NA> <NA> Asian <NA> <NA> <NA>
## 48658 SRP120018 GEO SAMN07789037 Homo sapiens TRANSCRIPTOMIC HepLPCs <NA> <NA> <NA> Asian <NA> <NA> <NA>
## 48659 SRP120018 GEO SAMN07789041 Homo sapiens TRANSCRIPTOMIC HepLPCs <NA> <NA> <NA> Asian <NA> <NA> <NA>
## clinical_information RACE health_state Population DONOR_HEALTH_STATUS DONOR_ETHNICITY reported_race ReleaseDate create_date ancestry donor_id Sample_type
## 48582 <NA> <NA> <NA> <NA> <NA> <NA> <NA> 2017-10-30 2017-10-24 11:32:00 <NA> <NA> <NA>
## 48583 <NA> <NA> <NA> <NA> <NA> <NA> <NA> 2017-10-30 2017-10-16 12:24:00 <NA> <NA> <NA>
## 48656 <NA> <NA> <NA> <NA> <NA> <NA> <NA> 2017-10-30 2017-10-16 12:11:00 <NA> <NA> <NA>
## 48657 <NA> <NA> <NA> <NA> <NA> <NA> <NA> 2017-10-30 2017-10-16 12:12:00 <NA> <NA> <NA>
## 48658 <NA> <NA> <NA> <NA> <NA> <NA> <NA> 2017-10-30 2017-10-16 12:10:00 <NA> <NA> <NA>
## 48659 <NA> <NA> <NA> <NA> <NA> <NA> <NA> 2017-10-30 2017-10-16 12:05:00 <NA> <NA> <NA>
## source_name source_material_cell_type source_material Organism_part race.ethnicity primary_race SAMPLE_TYPE Sampling_site finalRace finalGeography hispanic
## 48582 Donor_1 TEM P5 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> Asia <NA>
## 48583 Donor_2 PHCs <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> Asia <NA>
## 48656 Donor_1 HepLPCs-Hep <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> Asia <NA>
## 48657 Donor_1 TEM Day10 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> Asia <NA>
## 48658 Donor_2 TEM Day10 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> Asia <NA>
## 48659 Donor_2 TEM P5 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> Asia <NA>The reason I went down this rabbit hole was that there were about 3000 samples that didn’t have tissue info if we focused on the most complete categories: Cell_type and Tissue. I checked some manually and most of them seemed to have info in source_name and Organism_part, which I then rescued from the original SRA downloads… so let’s look at whether we really need the info from the other columns too:
head(allSRAFinal[is.na(allSRAFinal$Tissue) & is.na(allSRAFinal$Cell_type), ])## SRA.Study Center.Name BioSample Organism LibrarySource Cell_type Tissue Cell_line disease ETHNICITY disease_state Diagnosis tissue_type clinical_information
## 49372 SRP191569 GEO SAMN11372981 Homo sapiens TRANSCRIPTOMIC <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 49382 SRP191569 GEO SAMN11372986 Homo sapiens TRANSCRIPTOMIC <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 49383 SRP191569 GEO SAMN11372985 Homo sapiens TRANSCRIPTOMIC <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 49384 SRP191569 GEO SAMN11372984 Homo sapiens TRANSCRIPTOMIC <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 49385 SRP191569 GEO SAMN11372983 Homo sapiens TRANSCRIPTOMIC <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 49386 SRP191569 GEO SAMN11372980 Homo sapiens TRANSCRIPTOMIC <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## RACE health_state Population DONOR_HEALTH_STATUS DONOR_ETHNICITY reported_race ReleaseDate create_date ancestry donor_id Sample_type source_name
## 49372 African American <NA> <NA> <NA> <NA> <NA> 2019-08-24 2019-04-09 14:25:00 <NA> <NA> <NA> PBMCs
## 49382 African American <NA> <NA> <NA> <NA> <NA> 2019-08-24 2019-04-09 14:21:00 <NA> <NA> <NA> PBMCs
## 49383 African American <NA> <NA> <NA> <NA> <NA> 2019-08-24 2019-04-09 14:21:00 <NA> <NA> <NA> PBMCs
## 49384 African American <NA> <NA> <NA> <NA> <NA> 2019-08-24 2019-04-09 14:21:00 <NA> <NA> <NA> PBMCs
## 49385 African American <NA> <NA> <NA> <NA> <NA> 2019-08-24 2019-04-09 14:23:00 <NA> <NA> <NA> PBMCs
## 49386 African American <NA> <NA> <NA> <NA> <NA> 2019-08-24 2019-04-09 14:22:00 <NA> <NA> <NA> PBMCs
## source_material_cell_type source_material Organism_part race.ethnicity primary_race SAMPLE_TYPE Sampling_site finalRace finalGeography hispanic
## 49372 <NA> <NA> <NA> <NA> <NA> <NA> <NA> Black or African American <NA> <NA>
## 49382 <NA> <NA> <NA> <NA> <NA> <NA> <NA> Black or African American <NA> <NA>
## 49383 <NA> <NA> <NA> <NA> <NA> <NA> <NA> Black or African American <NA> <NA>
## 49384 <NA> <NA> <NA> <NA> <NA> <NA> <NA> Black or African American <NA> <NA>
## 49385 <NA> <NA> <NA> <NA> <NA> <NA> <NA> Black or African American <NA> <NA>
## 49386 <NA> <NA> <NA> <NA> <NA> <NA> <NA> Black or African American <NA> <NA>head(allSRAFinal[is.na(allSRAFinal$Tissue) & is.na(allSRAFinal$Cell_type) & is.na(allSRAFinal$source_name), ])## SRA.Study Center.Name BioSample Organism LibrarySource Cell_type Tissue Cell_line disease ETHNICITY disease_state
## 188456 ERP116722 Exeter Sequencing Service SAMEA5854572 Homo sapiens TRANSCRIPTOMIC <NA> <NA> <NA> cerebral malaria Mandingo <NA>
## 188457 ERP116722 Exeter Sequencing Service SAMEA5854573 Homo sapiens TRANSCRIPTOMIC <NA> <NA> <NA> cerebral malaria Jola <NA>
## 188459 ERP116722 Exeter Sequencing Service SAMEA5854574 Homo sapiens TRANSCRIPTOMIC <NA> <NA> <NA> cerebral malaria Mandingo <NA>
## 188460 ERP116722 Exeter Sequencing Service SAMEA5854575 Homo sapiens TRANSCRIPTOMIC <NA> <NA> <NA> cerebral malaria Fulla <NA>
## 188461 ERP116722 Exeter Sequencing Service SAMEA5854576 Homo sapiens TRANSCRIPTOMIC <NA> <NA> <NA> cerebral malaria Fulla <NA>
## 188463 ERP116722 Exeter Sequencing Service SAMEA5854577 Homo sapiens TRANSCRIPTOMIC <NA> <NA> <NA> cerebral malaria and hyperlactatemia Jola <NA>
## Diagnosis tissue_type clinical_information RACE health_state Population DONOR_HEALTH_STATUS DONOR_ETHNICITY reported_race ReleaseDate create_date ancestry donor_id
## 188456 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> 2019-08-07 2022-07-24 09:35:00 <NA> <NA>
## 188457 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> 2019-08-07 2021-10-10 16:52:00 <NA> <NA>
## 188459 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> 2019-08-07 2021-10-10 18:19:00 <NA> <NA>
## 188460 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> 2019-08-07 2022-07-18 16:45:00 <NA> <NA>
## 188461 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> 2019-08-07 2021-10-10 17:04:00 <NA> <NA>
## 188463 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> 2019-08-07 2022-07-15 18:32:00 <NA> <NA>
## Sample_type source_name source_material_cell_type source_material Organism_part race.ethnicity primary_race SAMPLE_TYPE Sampling_site finalRace finalGeography hispanic
## 188456 <NA> <NA> <NA> <NA> blood <NA> <NA> <NA> <NA> <NA> Subsaharan Africa <NA>
## 188457 <NA> <NA> <NA> <NA> blood <NA> <NA> <NA> <NA> <NA> Subsaharan Africa <NA>
## 188459 <NA> <NA> <NA> <NA> blood <NA> <NA> <NA> <NA> <NA> Subsaharan Africa <NA>
## 188460 <NA> <NA> <NA> <NA> blood <NA> <NA> <NA> <NA> <NA> Subsaharan Africa <NA>
## 188461 <NA> <NA> <NA> <NA> blood <NA> <NA> <NA> <NA> <NA> Subsaharan Africa <NA>
## 188463 <NA> <NA> <NA> <NA> blood <NA> <NA> <NA> <NA> <NA> Subsaharan Africa <NA>dim(allSRAFinal[is.na(allSRAFinal$Tissue) & is.na(allSRAFinal$Cell_type) & is.na(allSRAFinal$source_name) & is.na(allSRAFinal$tissue_type) & is.na(allSRAFinal$Organism_part), ])## [1] 464 36noInfo <- allSRAFinal[is.na(allSRAFinal$Tissue) & is.na(allSRAFinal$Cell_type) & is.na(allSRAFinal$source_name) & is.na(allSRAFinal$tissue_type) &is.na(allSRAFinal$Organism_part), ]
table(noInfo$SRA.Study) ##
## ERP001942
## 464by(noInfo, noInfo$SRA.Study, function(x) head(x, n=20)) ## noInfo$SRA.Study: ERP001942
## SRA.Study Center.Name BioSample Organism LibrarySource Cell_type Tissue Cell_line disease ETHNICITY disease_state Diagnosis tissue_type clinical_information RACE
## 254802 ERP001942 UNIGE SAMEA1573207 Homo sapiens TRANSCRIPTOMIC <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 254803 ERP001942 UNIGE SAMEA1573203 Homo sapiens TRANSCRIPTOMIC <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 254804 ERP001942 MPIMG SAMEA1573212 Homo sapiens TRANSCRIPTOMIC <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 254805 ERP001942 UNIGE SAMEA1573516 Homo sapiens TRANSCRIPTOMIC <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 254806 ERP001942 CRG SAMEA1573527 Homo sapiens TRANSCRIPTOMIC <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 254807 ERP001942 MPIMG SAMEA1573339 Homo sapiens TRANSCRIPTOMIC <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 254808 ERP001942 LUMC SAMEA1573507 Homo sapiens TRANSCRIPTOMIC <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 254809 ERP001942 UNIGE SAMEA1573345 Homo sapiens TRANSCRIPTOMIC <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 254810 ERP001942 MPIMG SAMEA1573368 Homo sapiens TRANSCRIPTOMIC <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 254811 ERP001942 ICMB SAMEA1573523 Homo sapiens TRANSCRIPTOMIC <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 254812 ERP001942 UNIGE SAMEA1573219 Homo sapiens TRANSCRIPTOMIC <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 254813 ERP001942 UNIGE SAMEA1573202 Homo sapiens TRANSCRIPTOMIC <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 254814 ERP001942 ICMB SAMEA1573188 Homo sapiens TRANSCRIPTOMIC <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 254815 ERP001942 HMGU SAMEA1573508 Homo sapiens TRANSCRIPTOMIC <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 254816 ERP001942 UNIGE SAMEA1573329 Homo sapiens TRANSCRIPTOMIC <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 254817 ERP001942 LUMC SAMEA1573101 Homo sapiens TRANSCRIPTOMIC <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 254818 ERP001942 ICMB SAMEA1573158 Homo sapiens TRANSCRIPTOMIC <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 254819 ERP001942 UNIGE SAMEA1573233 Homo sapiens TRANSCRIPTOMIC <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 254820 ERP001942 LUMC SAMEA1573467 Homo sapiens TRANSCRIPTOMIC <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 254821 ERP001942 UNIGE SAMEA1573324 Homo sapiens TRANSCRIPTOMIC <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## health_state Population DONOR_HEALTH_STATUS DONOR_ETHNICITY reported_race ReleaseDate create_date ancestry donor_id Sample_type source_name
## 254802 <NA> CEU <NA> <NA> <NA> 2012-11-07 2022-07-21 01:31:00 <NA> <NA> <NA> <NA>
## 254803 <NA> CEU <NA> <NA> <NA> 2012-11-07 2021-10-17 22:59:00 <NA> <NA> <NA> <NA>
## 254804 <NA> TSI <NA> <NA> <NA> 2012-11-07 2022-07-21 03:21:00 <NA> <NA> <NA> <NA>
## 254805 <NA> CEU <NA> <NA> <NA> 2012-11-07 2021-10-17 23:08:00 <NA> <NA> <NA> <NA>
## 254806 <NA> FIN <NA> <NA> <NA> 2012-11-07 2021-10-17 23:14:00 <NA> <NA> <NA> <NA>
## 254807 <NA> YRI <NA> <NA> <NA> 2012-11-07 2022-07-16 17:35:00 <NA> <NA> <NA> <NA>
## 254808 <NA> CEU <NA> <NA> <NA> 2012-11-07 2022-07-22 21:13:00 <NA> <NA> <NA> <NA>
## 254809 <NA> FIN <NA> <NA> <NA> 2012-11-07 2022-07-17 20:28:00 <NA> <NA> <NA> <NA>
## 254810 <NA> FIN <NA> <NA> <NA> 2012-11-07 2022-07-22 20:59:00 <NA> <NA> <NA> <NA>
## 254811 <NA> YRI <NA> <NA> <NA> 2012-11-07 2021-10-17 22:57:00 <NA> <NA> <NA> <NA>
## 254812 <NA> YRI <NA> <NA> <NA> 2012-11-07 2022-07-16 21:14:00 <NA> <NA> <NA> <NA>
## 254813 <NA> FIN <NA> <NA> <NA> 2012-11-07 2022-07-22 21:04:00 <NA> <NA> <NA> <NA>
## 254814 <NA> TSI <NA> <NA> <NA> 2012-11-07 2022-07-13 08:22:00 <NA> <NA> <NA> <NA>
## 254815 <NA> TSI <NA> <NA> <NA> 2012-11-07 2022-07-22 21:10:00 <NA> <NA> <NA> <NA>
## 254816 <NA> YRI <NA> <NA> <NA> 2012-11-07 2022-07-14 22:14:00 <NA> <NA> <NA> <NA>
## 254817 <NA> FIN <NA> <NA> <NA> 2012-11-07 2022-07-14 08:22:00 <NA> <NA> <NA> <NA>
## 254818 <NA> CEU <NA> <NA> <NA> 2012-11-07 2021-10-17 23:08:00 <NA> <NA> <NA> <NA>
## 254819 <NA> GBR <NA> <NA> <NA> 2012-11-07 2022-07-13 08:29:00 <NA> <NA> <NA> <NA>
## 254820 <NA> TSI <NA> <NA> <NA> 2012-11-07 2022-07-16 19:43:00 <NA> <NA> <NA> <NA>
## 254821 <NA> YRI <NA> <NA> <NA> 2012-11-07 2022-07-16 17:22:00 <NA> <NA> <NA> <NA>
## source_material_cell_type source_material Organism_part race.ethnicity primary_race SAMPLE_TYPE Sampling_site finalRace finalGeography hispanic
## 254802 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> Europe <NA>
## 254803 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> Europe <NA>
## 254804 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> Europe <NA>
## 254805 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> Europe <NA>
## 254806 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> Europe <NA>
## 254807 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> Subsaharan Africa <NA>
## 254808 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> Europe <NA>
## 254809 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> Europe <NA>
## 254810 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> Europe <NA>
## 254811 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> Subsaharan Africa <NA>
## 254812 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> Subsaharan Africa <NA>
## 254813 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> Europe <NA>
## 254814 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> Europe <NA>
## 254815 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> Europe <NA>
## 254816 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> Subsaharan Africa <NA>
## 254817 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> Europe <NA>
## 254818 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> Europe <NA>
## 254819 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> Europe <NA>
## 254820 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> Europe <NA>
## 254821 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> Subsaharan Africa <NA>The only one there is the LCLs, ERP001942, which we’ll fix later. Looking at the table there’s a lot of redundancy between columns, or a lot of studies that report different kinds of info in different columns, so it is worth being strategic so we’re not trying to disambiguate a bunch of weird things.
Once again, false/false is where we want to be looking
table(is.na(allSRAFinal$Tissue), is.na(allSRAFinal$Cell_type)) # Fair amount of overlap##
## FALSE TRUE
## FALSE 2568 13886
## TRUE 2333 6601table(is.na(allSRAFinal$Tissue), is.na(allSRAFinal$source_name)) # Lots of overlap##
## FALSE TRUE
## FALSE 7303 9151
## TRUE 7167 1767table(is.na(allSRAFinal$Tissue), is.na(allSRAFinal$Organism_part)) # No one uses both of these at the same time##
## FALSE TRUE
## FALSE 0 16454
## TRUE 467 8467table(is.na(allSRAFinal$Cell_type), is.na(allSRAFinal$source_name))##
## FALSE TRUE
## FALSE 2824 2077
## TRUE 11646 8841table(is.na(allSRAFinal$Cell_type), is.na(allSRAFinal$Organism_part))##
## FALSE TRUE
## FALSE 20 4881
## TRUE 447 20040source_name is clearly the most redundant one, but I grabbed all of them anyhow and did some manual spotchecking and cleaning up of the term list, which I think is mostly ok/ok enough that results won’t change too much. We’ll want to keep an eye out on the disease columns later on, though, and on the new disease column I made from those…
Now we explore a few ways of merging data… First we sanity check the order in which to merge…
# Merge the two tissue type columns:
allSRAFinal$twoTissue <- coalesce(allSRAFinal$Tissue, allSRAFinal$tissue_type)
length(table(allSRAFinal$twoTissue)) # A few less than when you sum the other two together, but that's expected.
table(is.na(allSRAFinal$twoTissue)) # 18k have some sort of tissue assignment
allSRAFinal$threeTissue <- coalesce(allSRAFinal$twoTissue, allSRAFinal$Organism_part)
length(table(allSRAFinal$threeTissue)) # A few less than when you sum the other two together, but that's expected.
table(is.na(allSRAFinal$threeTissue)) # only gained like 600...
allSRAFinal$fourTissue <- coalesce(allSRAFinal$threeTissue, allSRAFinal$Cell_type)
length(table(allSRAFinal$fourTissue)) # A few less than when you sum the other two together, but that's expected.
table(is.na(allSRAFinal$fourTissue)) # 21k have some sort of tissue assignment
allSRAFinal$fiveTissue <- coalesce(allSRAFinal$fourTissue, allSRAFinal$source_name)
length(table(allSRAFinal$fiveTissue)) # A few less than when you sum the other two together, but that's expected.
table(is.na(allSRAFinal$fiveTissue)) # 21k have some sort of tissue assignment
# The missing 464 are the LCLS from that one paper, so we're good there, we'll fix them later.
# What kind of gaps did we fill with each step?
head(allSRAFinal[!is.na(allSRAFinal$twoTissue) & is.na(allSRAFinal$Tissue),], n = 20)
head(allSRAFinal[!is.na(allSRAFinal$threeTissue) & is.na(allSRAFinal$twoTissue),], n = 20)
head(allSRAFinal[!is.na(allSRAFinal$fourTissue) & is.na(allSRAFinal$threeTissue),], n = 20)
head(allSRAFinal[!is.na(allSRAFinal$fiveTissue) & is.na(allSRAFinal$fourTissue),], n = 20)
head(sort(table(allSRAFinal$twoTissue), decreasing = T)) #
head(sort(table(allSRAFinal$threeTissue), decreasing = T)) #
head(sort(table(allSRAFinal$fourTissue), decreasing = T)) #
head(sort(table(allSRAFinal$fiveTissue), decreasing = T)) #And now it’s time to update each of those, just for sanity checking, before we pick which approach to go with
For some reason the match command won’t work well with the indexing inside the command, so we gotta separate it out. Don’t care to troubleshoot.
cleanInfo <- read.csv("20240616_cell_type_tissue_descriptors.csv")
cleanTissue <- cleanInfo[cleanInfo$coded.as %in% "tissue",]
cleanTissue_type <- cleanInfo[cleanInfo$coded.as %in% "tissue_type",]
cleanOrganism_part <- cleanInfo[cleanInfo$coded.as %in% "Organism_part",]
cleanCell_type <- cleanInfo[cleanInfo$coded.as %in% "cell_type",]
cleanSource_name <- cleanInfo[cleanInfo$coded.as %in% "source_name",]
# This is going to be really annoying, but I think it's easiest to do it by coalescing five different updated columns at the end... except it's ten, because we do system and organ.
allSRAFinal$tissueSystem <- cleanTissue[match(allSRAFinal$Tissue, cleanTissue$Term, incomparables = NA, nomatch = NA),]$system
allSRAFinal$tissue_typeSystem <- cleanTissue_type[match(allSRAFinal$tissue_type, cleanTissue_type$Term, incomparables = NA, nomatch = NA),]$system
allSRAFinal$Organism_partSystem <- cleanOrganism_part[match(allSRAFinal$Organism_part, cleanOrganism_part$Term, incomparables = NA, nomatch = NA),]$system
allSRAFinal$Cell_typeSystem <- cleanCell_type[match(allSRAFinal$Cell_type, cleanCell_type$Term, incomparables = NA, nomatch = NA),]$system
allSRAFinal$source_nameSystem <- cleanSource_name[match(allSRAFinal$source_name, cleanSource_name$Term, incomparables = NA, nomatch = NA),]$system
allSRAFinal$tissueOrgan <- cleanTissue[match(allSRAFinal$Tissue, cleanTissue$Term, incomparables = NA, nomatch = NA),]$organ
allSRAFinal$tissue_typeOrgan <- cleanTissue_type[match(allSRAFinal$tissue_type, cleanTissue_type$Term, incomparables = NA, nomatch = NA),]$organ
allSRAFinal$Organism_partOrgan <- cleanOrganism_part[match(allSRAFinal$Organism_part, cleanOrganism_part$Term, incomparables = NA, nomatch = NA),]$organ
allSRAFinal$Cell_typeOrgan <- cleanCell_type[match(allSRAFinal$Cell_type, cleanCell_type$Term, incomparables = NA, nomatch = NA),]$organ
allSRAFinal$source_nameOrgan <- cleanSource_name[match(allSRAFinal$source_name, cleanSource_name$Term, incomparables = NA, nomatch = NA),]$organ
# And now we coalesce in the order we established above...
allSRAFinal$finalSystem <- coalesce(allSRAFinal$tissueSystem, allSRAFinal$tissue_typeSystem, allSRAFinal$Organism_partSystem, allSRAFinal$Cell_typeSystem, allSRAFinal$source_nameSystem)
allSRAFinal$finalOrgan <- coalesce(allSRAFinal$tissueOrgan, allSRAFinal$tissue_typeOrgan, allSRAFinal$Organism_partOrgan, allSRAFinal$Cell_typeOrgan, allSRAFinal$source_nameOrgan)So that seems like it should have worked, but we should sanity check it a bit more… First, do things agree across the descriptor levels, and if not, what should we do?
table(allSRAFinal$finalSystem, allSRAFinal$tissueSystem) %>% melt (.) %>%
ggplot(., aes(Var.2, Var.1)) +
geom_tile(aes(fill = log2(value)), colour = "white") +
scale_fill_gradient(low = "white", high = "red") +
ggtitle("System annotation") +
xlab("Tissue") +
ylab("Final") +
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
table(allSRAFinal$finalSystem, allSRAFinal$tissue_typeSystem) %>% melt (.) %>%
ggplot(., aes(Var.2, Var.1)) +
geom_tile(aes(fill = log2(value)), colour = "white") +
scale_fill_gradient(low = "white", high = "red") +
ggtitle("System annotation") +
xlab("tissue_type") +
ylab("Final") +
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
table(allSRAFinal$finalSystem, allSRAFinal$Organism_partSystem) %>% melt (.) %>%
ggplot(., aes(Var.2, Var.1)) +
geom_tile(aes(fill = log2(value)), colour = "white") +
scale_fill_gradient(low = "white", high = "red") +
ggtitle("System annotation") +
xlab("Organism_part") +
ylab("Final") +
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
# Yeeeeeep these two are fucked up. Some of it will have been my old approach, but other stuff is gonna be harder to clean up.
table(allSRAFinal$finalSystem, allSRAFinal$Cell_typeSystem) %>% melt (.) %>%
ggplot(., aes(Var.2, Var.1)) +
geom_tile(aes(fill = log2(value)), colour = "white") +
scale_fill_gradient(low = "white", high = "red") +
ggtitle("System annotation") +
xlab("Cell_type") +
ylab("Final") +
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
table(allSRAFinal$finalSystem, allSRAFinal$source_nameSystem) %>% melt (.) %>%
ggplot(., aes(Var.2, Var.1)) +
geom_tile(aes(fill = log2(value)), colour = "white") +
scale_fill_gradient(low = "white", high = "red") +
ggtitle("System annotation") +
xlab("source_name") +
ylab("Final") +
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
# And now same for organs:
table(allSRAFinal$finalOrgan, allSRAFinal$tissueOrgan) %>% melt (.) %>%
ggplot(., aes(Var.2, Var.1)) +
geom_tile(aes(fill = log2(value)), colour = "white") +
scale_fill_gradient(low = "white", high = "red") +
ggtitle("Organ annotation") +
xlab("Tissue") +
ylab("Final") +
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
table(allSRAFinal$finalOrgan, allSRAFinal$tissue_typeOrgan) %>% melt (.) %>%
ggplot(., aes(Var.2, Var.1)) +
geom_tile(aes(fill = log2(value)), colour = "white") +
scale_fill_gradient(low = "white", high = "red") +
ggtitle("Organ annotation") +
xlab("tissue_type") +
ylab("Final") +
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
table(allSRAFinal$finalOrgan, allSRAFinal$Organism_partOrgan) %>% melt (.) %>%
ggplot(., aes(Var.2, Var.1)) +
geom_tile(aes(fill = log2(value)), colour = "white") +
scale_fill_gradient(low = "white", high = "red") +
ggtitle("Organ annotation") +
xlab("Organism_part") +
ylab("Final") +
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
# Yeeeeeep these two are fucked up. Some of it will have been my old approach, but other stuff is gonna be harder to clean up.
table(allSRAFinal$finalOrgan, allSRAFinal$Cell_typeOrgan) %>% melt (.) %>%
ggplot(., aes(Var.2, Var.1)) +
geom_tile(aes(fill = log2(value)), colour = "white") +
scale_fill_gradient(low = "white", high = "red") +
ggtitle("Organ annotation") +
xlab("Cell_type") +
ylab("Final") +
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
table(allSRAFinal$finalOrgan, allSRAFinal$source_nameOrgan) %>% melt (.) %>%
ggplot(., aes(Var.2, Var.1)) +
geom_tile(aes(fill = log2(value)), colour = "white") +
scale_fill_gradient(low = "white", high = "red") +
ggtitle("Organ annotation") +
xlab("source_name") +
ylab("Final") +
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
# How often do these two agree?
table(allSRAFinal$Cell_typeOrgan, allSRAFinal$source_nameOrgan) %>% melt (.) %>%
ggplot(., aes(Var.2, Var.1)) +
geom_tile(aes(fill = log2(value)), colour = "white") +
scale_fill_gradient(low = "white", high = "red") +
ggtitle("Organ annotation") +
xlab("source_name") +
ylab("Cell_type") +
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
# How often do these two agree?
table(allSRAFinal$finalOrgan, allSRAFinal$finalSystem) %>% melt (.) %>%
ggplot(., aes(Var.2, Var.1)) +
geom_tile(aes(fill = log2(value)), colour = "white") +
scale_fill_gradient(low = "white", high = "red") +
ggtitle("Master annotation") +
xlab("System") +
ylab("Organ") +
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
Uh oh… There’s a fair bit of disagreement when we compare the cell type based assignments and the tissue based assignments. And some kind of make sense, and some really don’t, like, intestine/iPSC or reproductive/digestive
# Let's look only at the discrepancies... write a little function to show us because otherwise it gets annoying
seeMismatches <- function(Var1, Var2){
messySubset <- allSRAFinal[allSRAFinal[[Var1]] != allSRAFinal[[Var2]],] %>% .[!is.na(.[[Var2]]),]
return(messySubset)
}
cellOrganMismatches <- seeMismatches("finalOrgan", "Cell_typeOrgan")
cellSystemMismatches <- seeMismatches("finalSystem", "Cell_typeSystem")
sourceOrganMismatches <- seeMismatches("finalOrgan", "source_nameOrgan")
sourceSystemMismatches <- seeMismatches("finalSystem", "source_nameSystem")
table(cellOrganMismatches$finalOrgan, cellOrganMismatches$Cell_typeOrgan)##
## blood cancer heart iPSC uterus
## blood vessel 0 0 1 0 0
## bone marrow 31 0 0 0 0
## intestine 0 0 0 39 0
## lung 66 0 12 0 0
## ovary 0 17 0 0 1
## skin 0 0 0 15 0table(cellSystemMismatches$finalSystem, cellSystemMismatches$Cell_typeSystem)##
## cancer cardiovascular connective tissue immune
## anatomical structure 0 0 12 0
## connective tissue 10 0 0 0
## hematopoetic 0 0 0 31
## integument 0 0 36 0
## reproductive 17 0 0 0
## respiratory 0 12 0 66table(sourceOrganMismatches$finalOrgan, sourceOrganMismatches$source_nameOrgan)##
## cancer lung lymph node
## blood 0 8 66
## ovary 9 0 0
## prostate 16 0 0
## uterus 4 0 0table(sourceSystemMismatches$finalSystem, sourceSystemMismatches$source_nameSystem)##
## cancer connective tissue lymphatic respiratory
## immune 0 0 66 8
## reproductive 29 129 0 0# Make my life easier
columnsILike <- c(1:3, 6,7,8,13,26,29,44,45,49,50,51,52)
# by(cellOrganMismatches, c(cellOrganMismatches$finalOrgan), function(x) head(x[,columnsILike], n = 20))
# by(cellSystemMismatches, c(cellSystemMismatches$Cell_typeSystem), function(x) head(x[,columnsILike], n = 20))
# by(sourceOrganMismatches, c(sourceOrganMismatches$finalOrgan), function(x) head(x[,columnsILike], n = 20))
# by(sourceSystemMismatches, c(sourceSystemMismatches$finalSystem), function(x) head(x[,columnsILike], n = 20))
table(cellOrganMismatches$SRA.Study)##
## DRP001797 SRP221289 SRP221290 SRP223239 SRP259538 SRP266877 SRP310133 SRP328303 SRP362734
## 1 8 10 66 12 15 20 11 39table(cellSystemMismatches$SRA.Study)##
## SRP070663 SRP125571 SRP219483 SRP220383 SRP221289 SRP221290 SRP223239 SRP259538 SRP310133 SRP328303
## 4 24 10 10 7 20 66 12 20 11table(sourceOrganMismatches$SRA.Study)##
## SRP118614 SRP223239 SRP290921 SRP352989
## 16 66 13 8table(sourceSystemMismatches$SRA.Study)##
## SRP118614 SRP223239 SRP225709 SRP290921 SRP352989
## 16 66 129 13 8# by(cellOrganMismatches, c(cellOrganMismatches$SRA.Study), function(x) head(x[,columnsILike], n = 20))
# by(cellSystemMismatches, c(cellSystemMismatches$SRA.Study), function(x) head(x[,columnsILike], n = 20))
# by(sourceOrganMismatches, c(sourceOrganMismatches$SRA.Study), function(x) head(x[,columnsILike], n = 20))
# by(sourceSystemMismatches, c(sourceSystemMismatches$SRA.Study), function(x) head(x[,columnsILike], n = 20))Some of these are clear incorrect assignments when cleaning up the tissue stuff, and so the big spreadsheet needs to be updated, eg, the foreskin fibroblasts from Coriell which came up as urogenital (SRP070663, now fixed). However, there’s also a lot of problems with samples that do end up in different classes depending of which column one looks at, eg, SRP266877 should really be under iPSC but isn’t because tissue is given as blood, even though it’s an established cell line… Not all the cases are clear cut, and some are straight up contradictory, eg, DRP001797, which says endocardial cells but also endothelium of blood vessels, or, SRP056287, which gives cell type as cortical neurons, and tissue as fibroblast and which seem to be iPSC-derived cortical neurons (updated), or SRP259538, which lists cardiomyocyte and foetal lung (but they’re cell-line derived anyhow)
There’s a lot of immune cells sampled in the lung, and I’m not sure what to do with them - keep them as both things?
So that’s a first pass, but we need to get rid of established cell lines and then come back to some of this. Next, some checks for missingness etc
table(allSRAFinal$finalSystem)##
## anatomical structure cancer cardiovascular connective tissue developmental digestive endocrine enteric
## 772 627 996 227 63 1313 56 12
## hematopoetic hepatic immune integument lymphatic muscular nervous reproductive
## 353 344 13621 408 40 823 1783 736
## respiratory skeletal urogenital visual
## 708 130 175 47table(allSRAFinal$finalOrgan)##
## adipose adrenal gland biliary tract bladder blastoderm blood blood vessel bone bone marrow brain breast
## 107 3 10 38 22 13618 222 35 353 1688 658
## cancer cartilage CNS digestive tract eye fallopian tube gallbladder heart intestine iPSC joint
## 637 3 79 59 47 8 3 774 1036 1457 92
## kidney larynx liver lung lymph node morula muscle neck nose oral cavity ovary
## 59 1 344 594 33 41 823 4 114 97 321
## pancreas pituitary gland placenta PNS prostate skin spleen stomach testis thymus thyroid
## 98 7 32 28 259 408 3 107 16 1 46
## tonsil trachea urinary tract uterus vagina
## 6 12 78 84 16table(is.na(allSRAFinal$finalSystem), is.na(allSRAFinal$finalOrgan))##
## FALSE TRUE
## FALSE 23128 106
## TRUE 1453 701table(is.na(allSRAFinal$finalSystem)) ##
## FALSE TRUE
## 23234 2154table(is.na(allSRAFinal$finalOrgan))##
## FALSE TRUE
## 24581 807noAssignment <- allSRAFinal[is.na(allSRAFinal$finalOrgan) & is.na(allSRAFinal$finalSystem),]
table(noAssignment$SRA.Study)##
## DRP001150 ERP001942 ERP122083 ERP122103 ERP122256 SRP151215 SRP154973 SRP179998 SRP181079 SRP181649 SRP186687 SRP343308 SRP344328
## 1 464 6 1 5 25 4 15 8 15 19 124 14# by(noAssignment, noAssignment$SRA.Study, function(x) head(x))Most of these had info in one of the tissue columns, but sometimes I had to go back to run selector and dig a bit deeper.
- DRP001150: cell mixture - tissue sample (yikes): no fix, will get dropped
- ERP001942: LCLs
- ERP122083: more cancer: finalOrgan:cancer and finalTissue:cancer
- ERP122103: single cancer sample: finalOrgan:cancer and finalTissue:cancer
- ERP122256: soft tissue cancer: finalOrgan:cancer and finalTissue:cancer
- SRP151215: LCLs
- SRP154973: ‘adjacent normal’ to cancers, lung cell carcinoma: finalOrgan:lung, finalSystem:respiratory
- SRP179998: brain sample that for some reason isn’t matching well: finalOrgan:brain and finalSystem:nervous
- SRP181079: BCR sequencing of B-cells; finalOrgan:blood and finalSystem:immune
- SRP181649: single established cell line: no fix, will get dropped
- SRP186687: established cancer cell lines: no fix, will get dropped
- SRP343308: Intestinal biopsies from multiple tissues: finalOrgan:intestine and finalSystem:digestive
- SRP344328: Intrahepatic cholangiocarcinoma: finalOrgan: liver and finalSystem:hepatic
# What are the NAs...? 464 Of them are the LCL study, so let's fix that, first of all:
allSRAFinal[allSRAFinal$SRA.Study %in% c("ERP001942", "SRP151215", "SRP181079"),]$finalSystem <- "immune"
allSRAFinal[allSRAFinal$SRA.Study %in% c("ERP001942", "SRP151215", "SRP181079"),]$finalOrgan <- "blood"
allSRAFinal[allSRAFinal$SRA.Study %in% c("ERP122083", "ERP122103", "ERP122256"),]$finalSystem <- "cancer"
allSRAFinal[allSRAFinal$SRA.Study %in% c("ERP122083", "ERP122103", "ERP122256"),]$finalOrgan <- "cancer"
allSRAFinal[allSRAFinal$SRA.Study %in% "SRP344328",]$finalSystem <- "hepatic"
allSRAFinal[allSRAFinal$SRA.Study %in% "SRP344328",]$finalOrgan <- "liver"
allSRAFinal[allSRAFinal$SRA.Study %in% "SRP343308",]$finalSystem <- "digestive"
allSRAFinal[allSRAFinal$SRA.Study %in% "SRP343308",]$finalOrgan <- "intestine"
allSRAFinal[allSRAFinal$SRA.Study %in% "SRP179998",]$finalSystem <- "nervous"
allSRAFinal[allSRAFinal$SRA.Study %in% "SRP179998",]$finalOrgan <- "brain"
allSRAFinal[allSRAFinal$SRA.Study %in% "SRP154973",]$finalSystem <- "respiratory"
allSRAFinal[allSRAFinal$SRA.Study %in% "SRP154973",]$finalOrgan <- "lung"
noAssignment <- allSRAFinal[is.na(allSRAFinal$finalOrgan) & is.na(allSRAFinal$finalSystem),]
table(noAssignment$SRA.Study)##
## DRP001150 SRP181649 SRP186687
## 1 15 19# by(noAssignment, noAssignment$SRA.Study, function(x) head(x))
# Remove that mixed tissue one.
allSRAFinal <- allSRAFinal[!(allSRAFinal$SRA.Study %in% "DRP001150"),]
rm(noAssignment)Much better… so we’ve resolved the NAs, but not the tissue ambiguities, which I think might just have to stay as is.
Now we look at the rows with info encoded in the Cell_line column, in addition to our organ and tissue assignments. Lots of these are cell lines that have ancestry info, but are established/immortalised, which was not included in our initial search… but should it have been? A key culprit is SRP186687, which is almost 1000 cancer samples with sequencing.
The brutal approach is to simply remove the entire study, and then come back again. But it would be smarter to also whitelist some terms and not just remove the column, or remove any rows with info in the column. SRP347253 has some epic cell line names, for instance, and all the occurrences of LCL etc should be kept - but first gotta check they’re not already covered in other columns!
table(!is.na(allSRAFinal$Cell_line))##
## FALSE TRUE
## 22966 2421allCellLine <- allSRAFinal[!is.na(allSRAFinal$Cell_line),]
sort(table(allCellLine$SRA.Study), decreasing=T)##
## SRP186687 SRP347253 ERP114122 SRP344545 SRP132693 SRP154009 SRP323748 SRP362734 SRP106527 SRP217300 SRP090531 SRP125571 SRP151215 SRP103111 SRP174206 SRP223674 SRP330940 SRP221374
## 979 304 159 108 91 44 40 39 30 30 29 28 25 24 24 24 24 23
## SRP075592 SRP237644 SRP408226 SRP292295 SRP191042 SRP222944 SRP292536 SRP056109 SRP073347 SRP367525 SRP266877 SRP269147 SRP155367 SRP178235 SRP179498 SRP193979 SRP259538 SRP334792
## 21 21 20 19 18 18 18 16 16 16 15 14 12 12 12 12 12 12
## SRP352928 SRP131505 SRP198410 SRP219483 SRP220383 SRP091668 SRP041840 SRP057446 SRP060253 SRP194087 SRP220267 SRP070663 SRP052896 SRP066625 SRP181649 SRP298738 SRP074707 SRP111915
## 12 10 10 10 10 8 6 6 6 6 6 4 3 3 3 3 2 2
## SRP164930
## 2head(sort(table(allCellLine$Cell_line), decreasing=T), n =50) # Some of these are clear keepers, some of the are clear removal. But the question is which is which, and how do I make my life easier? Let's check the big ones first:##
## lymphoblastoid cell line organoid raisin AGS
## 159 39 36 26
## PBMC HBMEC lymphoblastoid cell lines (LCLs) MCF7
## 26 25 25 25
## HepG2 motor neurons PC-3 (ATCC\\, CRL-1435) B lymphocyte
## 24 24 24 23
## SKOV-3 MDA-MB-231 PANC-1 primary vaginal epithelial cells
## 21 19 17 16
## WTC-11 merlot butter sun
## 15 14 13 13
## HB-8065 (Hep G2) [HEPG2] hiPSC-derived trophoblast IMR90-derived cardiomyocytes primarily peripheral blood mononuclear cells
## 12 12 12 12
## WC-24-02-DS-B WC-24-02-DS-M dijon lilac
## 12 12 11 11
## violet A549 corn EFO-21
## 11 10 10 10
## fibroblast grape HDLM-2 MDA-MB-468
## 10 10 10 10
## orchid 23555 28815 apple
## 10 9 9 9
## fibroblasts honey RWPE-1 21792
## 9 9 9 8
## 27322 29089 coffee H21792
## 8 8 8 8
## H25237 H28815
## 8 8# by(allCellLine, allCellLine$SRA.Study, function(x) sort(table(x$finalOrgan), decreasing=T))
# by(allCellLine, allCellLine$SRA.Study, function(x) sort(table(x$finalSystem), decreasing=T))
# by(allCellLine, allCellLine$SRA.Study, function(x) sort(table(x$Cell_line), decreasing=T))
# by(allCellLine, allCellLine$SRA.Study, function(x) head(x[,columnsILike], n=20))
rm(allCellLine)Let’s organise this a bit:
Studies to exclude:
- SRP186687: the huge cancer study, currently assigned to various organs and lots of NA. exclude.
- SRP217300: MIA PaCa, established breast cancer cell line. exclude.
- SRP090531: mix of established cell lines and lines that closely match other cellosaurus IDs. exclude.
- SRP103111: PC-3. established. exclude.
- SRP075592: established. exclude.
- SRP237644: established. exclude.
- SRP408226: HepG2. exclude.
- SRP292295: established cell line. exclude.
- SRP191042: established. exclude.
- SRP292536: established cancer cell lines. exclude.
- SRP056109: established cancer cell line. exclude.
- SRP269147: established cancer cell lines. exclude.
- SRP178235: established. exclude.
- SRP179498: established. exclude.
- SRP193979: HepG2. exclude.
- SRP259538: established cell line. exclude.
- SRP334792: prostate cancer cell lines from ATCC. exclude.
- SRP131505: established cancer cell lines. exclude.
- SRP091668: established. exclude.
- SRP041840: established. exclude.
- SRP057446: HepG2 and related. exclude.
- SRP060253: AGS. exclude.
- SRP194087: established. exclude.
- SRP220267: Caco2. exclude.
- SRP066625: HL-60/S4. exclude
- SRP181649: Huh-7. exclude.
- SRP298738: established AML cell line. exclude.
- SRP074707: established. exclude.
- SRP111915: established. exclude.
- SRP164930: established. exclude.
Studies to keep:
These are mostly iPSCs and derivatives, so not too sure what to do with them:
- SRP347253: the big iPSC panel, correctly assigned to iPSCs. ok
- SRP132693: iPSC, iPSC derived cardiomyocytes and primary heart tissue. correctly labelled. ok.
- SRP154009: iPSC and iPSC derived cardiomyocytes. correctly labelled. ok
- SRP323748: SRA accession says they’re all iPSC derived cardiomyocytes from iPSCore, and this is recorded in the ‘cell_subtype’ column. need to update. Unsure.
- SRP362734: colorectal cancer organoids… currently processed as if colorectal. need to update. unsure.
- SRP125571: iPSC-derived cortical neurons (in
Sample_type) from coriell fibroblasts. need to update. unsure. - SRP223674: iPSC-derived motor neurons, incorrectly labelled as blood (starting material). need to update. unsure.
- SRP330940: iPSCs and iPSC-derived material from publicly available line. unsure.
- SRP222944: iPSC-derived tissues, labelled as peripheral blood. need to update. unsure.
- SRP266877: WTC-11 iPSC cell line. need to update. unsure.
- SRP155367: iPSC and iPSC derived material. SRA says from Coriell, but neither cellosaurus nor coriell have heard of this. unsure.
- SRP070663: Coriell cell lines and iPSCs from them, correctly assigned. unsure.
These are easy to justify keeping:
- ERP114122: LCLs, (from 1000 Genomes). ok
- SRP344545: LCLs, (from 1000 Genomes). ok
- SRP106527: CEU LCL sequencing. ok
- SRP151215: LCLs. ok.
- SRP174206: primary samples, contain solid info in tissue column (tho I’m still sus but…). ok.
- SRP221374: blood cells. ok.
- SRP073347: PBMCs. ok.
- SRP367525: primary vaginal samples. ok.
- SRP352928: PBMCs. ok.
- SRP198410: PBMCs. ok.
- SRP219483: primary cell lines. ok.
- SRP220383: primary cell lines. ok.
- SRP052896: primary samples. ok.
So that’s a quick summary. Let’s clean this up and then check out what happens with all of those disagreements from before; this has also identified a few that need to be updated:
# First we make a list of a lot of studies to exclude:
establishedStudies <- c("SRP164930", "SRP111915", "SRP074707", "SRP298738", "SRP181649", "SRP066625", "SRP220267", "SRP194087", "SRP060253", "SRP057446", "SRP041840", "SRP091668", "SRP131505", "SRP334792", "SRP259538", "SRP193979", "SRP179498", "SRP178235", "SRP269147", "SRP056109", "SRP292536", "SRP191042", "SRP292295", "SRP408226", "SRP237644", "SRP075592", "SRP103111", "SRP090531", "SRP217300", "SRP186687")
allSRAFinal <- allSRAFinal[!(allSRAFinal$SRA.Study %in% establishedStudies),]
dim(allSRAFinal) # Looks better## [1] 24028 48# Then we gotta make sure all the other tissue assignments are ok:
allCellLine <- allSRAFinal[allSRAFinal$SRA.Study %in% allSRAFinal[!is.na(allSRAFinal$Cell_line),]$SRA.Study,] # Yes again
by(allCellLine, allCellLine$SRA.Study, function(x) sort(table(x$finalOrgan), decreasing=T))## allCellLine$SRA.Study: ERP114122
## blood
## 159
## ---------------------------------------------------------------------------------------------------------------------------------------
## allCellLine$SRA.Study: SRP052896
## lung
## 1
## ---------------------------------------------------------------------------------------------------------------------------------------
## allCellLine$SRA.Study: SRP070663
##
## iPSC skin
## 8 4
## ---------------------------------------------------------------------------------------------------------------------------------------
## allCellLine$SRA.Study: SRP073347
## blood
## 16
## ---------------------------------------------------------------------------------------------------------------------------------------
## allCellLine$SRA.Study: SRP106527
## blood
## 30
## ---------------------------------------------------------------------------------------------------------------------------------------
## allCellLine$SRA.Study: SRP125571
##
## oral cavity skin
## 12 12
## ---------------------------------------------------------------------------------------------------------------------------------------
## allCellLine$SRA.Study: SRP132693
##
## iPSC heart
## 69 22
## ---------------------------------------------------------------------------------------------------------------------------------------
## allCellLine$SRA.Study: SRP151215
## blood
## 25
## ---------------------------------------------------------------------------------------------------------------------------------------
## allCellLine$SRA.Study: SRP154009
## iPSC
## 44
## ---------------------------------------------------------------------------------------------------------------------------------------
## allCellLine$SRA.Study: SRP155367
## iPSC
## 12
## ---------------------------------------------------------------------------------------------------------------------------------------
## allCellLine$SRA.Study: SRP174206
##
## blood skin
## 14 10
## ---------------------------------------------------------------------------------------------------------------------------------------
## allCellLine$SRA.Study: SRP198410
## blood
## 10
## ---------------------------------------------------------------------------------------------------------------------------------------
## allCellLine$SRA.Study: SRP219483
## skin
## 10
## ---------------------------------------------------------------------------------------------------------------------------------------
## allCellLine$SRA.Study: SRP220383
## skin
## 10
## ---------------------------------------------------------------------------------------------------------------------------------------
## allCellLine$SRA.Study: SRP221374
## blood
## 23
## ---------------------------------------------------------------------------------------------------------------------------------------
## allCellLine$SRA.Study: SRP222944
## blood
## 18
## ---------------------------------------------------------------------------------------------------------------------------------------
## allCellLine$SRA.Study: SRP223674
##
## blood skin
## 18 6
## ---------------------------------------------------------------------------------------------------------------------------------------
## allCellLine$SRA.Study: SRP266877
## skin
## 15
## ---------------------------------------------------------------------------------------------------------------------------------------
## allCellLine$SRA.Study: SRP323748
##
## skin blood
## 36 4
## ---------------------------------------------------------------------------------------------------------------------------------------
## allCellLine$SRA.Study: SRP330940
## iPSC
## 24
## ---------------------------------------------------------------------------------------------------------------------------------------
## allCellLine$SRA.Study: SRP344545
## blood
## 108
## ---------------------------------------------------------------------------------------------------------------------------------------
## allCellLine$SRA.Study: SRP347253
## iPSC
## 304
## ---------------------------------------------------------------------------------------------------------------------------------------
## allCellLine$SRA.Study: SRP352928
## blood
## 12
## ---------------------------------------------------------------------------------------------------------------------------------------
## allCellLine$SRA.Study: SRP362734
## intestine
## 39
## ---------------------------------------------------------------------------------------------------------------------------------------
## allCellLine$SRA.Study: SRP367525
## vagina
## 16# Some more updates...
allSRAFinal[allSRAFinal$SRA.Study %in% c("SRP362734", "SRP323748", "SRP266877", "SRP223674", "SRP222944", "SRP125571"),]$finalSystem <- NA
allSRAFinal[allSRAFinal$SRA.Study %in% c("SRP362734", "SRP323748", "SRP266877", "SRP223674", "SRP222944", "SRP125571"),]$finalOrgan <- "iPSC"
# I guess we could make all the plots again... but easier to just tally disagreements?
table(allSRAFinal$finalSystem, allSRAFinal$Cell_typeSystem) %>% melt (.) %>%
ggplot(., aes(Var.2, Var.1)) +
geom_tile(aes(fill = log2(value)), colour = "white") +
scale_fill_gradient(low = "white", high = "red") +
ggtitle("System annotation") +
xlab("Cell_type") +
ylab("Final") +
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
table(allSRAFinal$finalSystem, allSRAFinal$source_nameSystem) %>% melt (.) %>%
ggplot(., aes(Var.2, Var.1)) +
geom_tile(aes(fill = log2(value)), colour = "white") +
scale_fill_gradient(low = "white", high = "red") +
ggtitle("System annotation") +
xlab("source_name") +
ylab("Final") +
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
# And now same for organs:
table(allSRAFinal$finalOrgan, allSRAFinal$Cell_typeOrgan) %>% melt (.) %>%
ggplot(., aes(Var.2, Var.1)) +
geom_tile(aes(fill = log2(value)), colour = "white") +
scale_fill_gradient(low = "white", high = "red") +
ggtitle("Organ annotation") +
xlab("Cell_type") +
ylab("Final") +
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
table(allSRAFinal$finalOrgan, allSRAFinal$source_nameOrgan) %>% melt (.) %>%
ggplot(., aes(Var.2, Var.1)) +
geom_tile(aes(fill = log2(value)), colour = "white") +
scale_fill_gradient(low = "white", high = "red") +
ggtitle("Organ annotation") +
xlab("source_name") +
ylab("Final") +
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
# And the final table... this should be messy, but there should be some sanity to it.
table(allSRAFinal$finalOrgan, allSRAFinal$finalSystem) %>% melt(.) %>%
ggplot(., aes(Var.2, Var.1)) +
geom_tile(aes(fill = log2(value)), colour = "white") +
scale_fill_gradient(low = "white", high = "red") +
ggtitle("Master annotation") +
xlab("System") +
ylab("Organ") +
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
## Warning in type.convert.default(X[[i]], ...): 'as.is' should be specified by the caller; using TRUE
I think it all looks mostly good now, with the open question of what to do with those where there’s a mismatch between annotation derived from either Cell_type or source_name and the other columns. It’s not going to make a big difference, I guess, so I think we’re ready to move on to real analyses… let’s just look at things again tho:
dim(allSRAFinal)## [1] 24028 48cellOrganMismatches <- seeMismatches("finalOrgan", "Cell_typeOrgan")
cellSystemMismatches <- seeMismatches("finalSystem", "Cell_typeSystem")
sourceOrganMismatches <- seeMismatches("finalOrgan", "source_nameOrgan")
sourceSystemMismatches <- seeMismatches("finalSystem", "source_nameSystem")
table(cellOrganMismatches$finalOrgan, cellOrganMismatches$Cell_typeOrgan)##
## blood cancer heart uterus
## blood vessel 0 0 1 0
## bone marrow 31 0 0 0
## lung 66 0 0 0
## ovary 0 17 0 1table(cellSystemMismatches$finalSystem, cellSystemMismatches$Cell_typeSystem)##
## cancer connective tissue immune
## connective tissue 10 0 0
## hematopoetic 0 0 31
## integument 0 24 0
## reproductive 17 0 0
## respiratory 0 0 66table(sourceOrganMismatches$finalOrgan, sourceOrganMismatches$source_nameOrgan)##
## cancer lung lymph node
## blood 0 8 66
## lung 4 0 0
## ovary 9 0 0
## prostate 16 0 0
## uterus 4 0 0table(sourceSystemMismatches$finalSystem, sourceSystemMismatches$source_nameSystem)##
## cancer connective tissue lymphatic respiratory
## immune 0 0 66 8
## reproductive 29 129 0 0
## respiratory 4 0 0 0table(cellOrganMismatches$SRA.Study)##
## DRP001797 SRP221289 SRP221290 SRP223239 SRP310133 SRP328303
## 1 8 10 66 20 11table(cellSystemMismatches$SRA.Study)##
## SRP070663 SRP219483 SRP220383 SRP221289 SRP221290 SRP223239 SRP310133 SRP328303
## 4 10 10 7 20 66 20 11table(sourceOrganMismatches$SRA.Study)##
## SRP118614 SRP154973 SRP223239 SRP290921 SRP352989
## 16 4 66 13 8table(sourceSystemMismatches$SRA.Study)##
## SRP118614 SRP154973 SRP223239 SRP225709 SRP290921 SRP352989
## 16 4 66 129 13 8# Yep, not too fussed. So now we can remove some of the temp columns and write this out
allSRAFinal <- allSRAFinal[,-(37:46)]
saveRDS(allSRAFinal, "allSRAFinalTissues.rds")And we’re done!